Act III Close: Unnatural in Our Loops

Working notes, not prose. This page is a research packet for drafting the chapter: the brief, source material, questions, examples, reader perspectives, and the earlier draft.

Brief

Must: Show that we run all of these loops (learning, creating, becoming) with far more sophistication and speed than any other life form, across individuals, cultures and generations.
Serves: It gives the third sense in which we are unnatural, and hands the reader to the epilogue.

Research

Core argument

Start from sameness, not superiority. Every living thing runs loops: sense, respond, keep what works. A bacterium climbs a sugar gradient, a crow tries a twig, a population shifts over generations. Darwin’s “degree and not of kind” is the honest starting line. The chapter’s claim has to be about how far and how fast the loops go, not about a new ingredient only we have.
Name the three loops again, briefly, as they occur in other life. Learning (Ch. 14): many animals measure and adjust. Creating (Ch. 15): nest-builders, beavers and termites reshape matter (niche construction). Becoming (Ch. 16): play, exploration and idle attention show up in mammals and birds. None of the three is uniquely human, and the chapter should say so plainly.
Show the first difference: our loops ratchet. Animal traditions exist (39 chimpanzee behaviours, whale songs, sweet-potato washing), but they mostly stay inside what one clever individual could reinvent (the “zone of latent solutions”). Human culture keeps improvements and builds on them, with little slipping back. The loop’s output becomes the next loop’s starting point.
Show the second difference: our loops run across people, not just inside them. No one person could survive with only what they worked out alone (Franklin’s crew starved where the Netsilik thrived). Our learning is spread across teachers, tools, records and institutions. The loop that learns is a group, and often a group of the dead.
Show the third difference: our loops run across generations and media. Writing, number, recipes, drawings, code and law let a correction made in 1850 reach a child in 2026. That is Act II’s representation loop (Ch. 6) turned into a machine for inheritance. Genes, behaviour and symbols are separate channels of inheritance, and we are the species that leans hardest on the third.
Show the fourth difference: speed. Compare timescales. It took E. coli over 30,000 generations to evolve citrate use, and handaxes stayed much the same for over a million years. A new vaccine design moved from a posted genome sequence to authorized doses in under a year. Loops that used to take geological time now take one news cycle.
Show the fifth difference: loops about loops. We change how we learn, not just what we learn. Science is a loop built to catch errors in other loops. Double-loop learning (Argyris), peer review, the experimental method and a book like this one are examples. Hofstadter’s strange loop (Ch. 18) is the same move inside one head: the self is a loop that models itself.
Be honest about the cost. Sophisticated loops can run away (Ch. 18). They can carry a false belief as faithfully as a true one (Ch. 7, Ch. 19), and they can be lost when the carriers die (Tasmania, the Polar Inuit). The ratchet can slip. Speed without care is how you get arms races and outrage cycles. Being “unnatural in our loops” describes our reach. It does not certify our wisdom.
Gather the three unnaturals. Act I: we want in a way that looks close to free (volition). Act II: we can cut reality anywhere and move the pieces around in representation (distinction). Act III: we can run learning, creating and becoming across individuals, cultures and centuries, faster than anything else alive (loops). Each depends on the one before it: a loop needs a wanter, and a symbolic loop needs cuts to carry.
Hand off to the epilogue. If our loops reach this far, what we feed into them matters more than for any other creature. That sets up “Use the Blade, Build the World” without giving the practical advice again (that belongs to Ch. 22).

Key questions

If a chimpanzee can learn to crack nuts from its mother, what exactly is left for “unnatural” to mean? Is the difference one of degree, kind, or scale that becomes kind?
(Child’s question) Why don’t monkeys have schools? Why doesn’t a monkey grandmother tell stories about how she found the best tree?
Why did handaxes stay nearly unchanged for more than a million years, when phones change every year? What changed in the loop?
What does it mean for a loop to “ratchet”? What stops it slipping back, and what happens when that stops working?
(Cynic’s question) Isn’t “we are unnatural” just the old human exceptionalism in new clothes? Every species is unique at something. Bats echolocate and ants farm.
Could any single human, raised alone on an island, rebuild even a bow and arrow, never mind a pencil? If not, in what sense is human intelligence “ours”?
How much of what you know did you work out yourself, and how much did you inherit? Could you even tell the difference?
If the Franklin expedition had the best equipment and education of its day and still starved, what kind of knowledge were they missing? Who held it?
Is a library a memory? Is a recipe a loop? Where does the loop “live”: in brains, in books, or in between?
Can a culture forget? What did the Tasmanians and the Polar Inuit lose, and what does that say about how fragile our ratchet is?
If people improve a technology without understanding it (the Derex wheel experiment), who is doing the learning? The individuals, or the chain?
Do faster loops make us wiser, or only faster? Is there a speed at which a loop stops being able to correct itself?
(Cynic) Social media is a very fast, very sophisticated cultural loop. Is it evidence of our unnatural brilliance, or of how loops run away?
Is science a new kind of loop, or just ordinary learning with better bookkeeping?
Is “becoming” (the third loop) really sped up by culture, or is it the one loop that refuses to go faster? Can a generation inherit wonder the way it inherits arithmetic?
(Child) If I learn something today, does it help people who aren’t born yet? How?
What would it take for another species (or an AI) to join us in this third kind of unnaturalness? Would we recognise it?
How do the three unnaturals depend on each other? Could you have sophisticated loops without volition, or without the blade of distinction?
If our loops can change the world faster than we can measure the effects, what do we owe the people downstream of our loops?
Is it good news or bad news that we are unnatural in this way?

Examples

Bacterial learning versus bacterial evolution (cells). In Richard Lenski’s long-term E. coli experiment, twelve identical populations were started in 1988. None could use the citrate in their flask for over 30,000 generations, despite “billions of mutations”. One population finally managed it by about generation 31,500. Compare that with a person reading the word “citrate” on a label. Blount, Borland & Lenski 2008, PNAS
Imo and the sweet potatoes (animals, innovation spreading). On Kōjima (Koshima) island in Japan in the early 1950s, researchers left sweet potatoes on the beach for a macaque troop. A young female they named Imo began washing the sand off. The habit spread first to her peers and her mother, and older males largely never took it up. It is real animal culture, but it plateaued. It is also the source of the exaggerated “hundredth monkey” myth, a nice aside on how human loops distort stories. Wikipedia: Hundredth monkey effect
Chimpanzee cultures (animals, many traditions). In 1999 Whiten, Goodall and colleagues pooled 151 years of observation from seven field sites and found 39 behaviours (tool use, grooming, courtship) that are customary in some communities and absent in others, with no ecological reason for the difference. That is genuine culture, yet chimp nut-cracking hasn’t turned into mills. Whiten et al. 1999, Nature
Humpback song ripples (animals, horizontal transmission across an ocean). Over 11 years, new song types spread eastward across South Pacific humpback populations “like cultural ripples”. Animals can pass culture laterally at huge scales. The difference is that the songs replace each other rather than piling up. Garland et al. 2011, Current Biology
Handaxe stasis (early humans, deep time). Stone tools go back at least 2.6 million years. The Acheulean handaxe appeared about 1.76 million years ago and stayed in use, with only modest change, until roughly 400,000 to 250,000 years ago. That is well over a million years for one design. It is a useful contrast with the ratchet of the last 50,000 years. Smithsonian Human Origins
The Franklin expedition (people, the lost-explorer experiment). In 1845 Franklin’s two ships, “the best-equipped expedition in the history of British polar exploration” with a three-year food supply, got trapped off King William Island. Everyone died. The Netsilik had lived there for almost a millennium, and the main harbour’s name means “lots of fat”. Intelligence without inherited local knowledge was not enough. Burke and Wills (Australia, 1861) is a parallel case. Boyd, Richerson & Henrich 2011, PNAS
The Polar Inuit forget and relearn (groups, the ratchet slipping). An epidemic in the 1820s killed the older, knowledgeable members of the isolated Polar Inuit of northwest Greenland. By custom their possessions were buried with them. The group then lived without kayaks, bows and arrows, leisters and heat-saving snowhouse entrances until migrants from Baffin Island arrived around 1862. They adopted the tools at once. Boyd, Richerson & Henrich 2011. A related case is Tasmania’s toolkit, which lost complexity after the island was cut off from the mainland (Henrich 2004, American Antiquity 69:197–218).
The wheel that got better without anyone understanding it (lab, cumulative culture). In Derex et al.’s 2019 study, chains of participants tuned a weighted wheel rolling down a track. Each person saw the previous attempt. The wheels got faster over generations, but participants’ written theories stayed simple and often wrong, and passing on those theories narrowed exploration. Derex et al. 2019, Nature Human Behaviour
Languages that design themselves (lab, language as a loop). Kirby, Cornish and Smith (2008) passed an artificial language from learner to learner. It became more structured and easier to learn with no one intending it. This is the loop doing the design. Kirby et al. 2008, PNAS
Paper airplanes in 70 micro-societies (lab, how much copying is needed). Caldwell and Millen (2009) had 700 people build paper airplanes in chains. Imitation, emulation and teaching were each enough on their own for designs to improve cumulatively. This is useful for the “what exactly is special?” question. Caldwell & Millen 2009, Psychological Science
Lactose and dairying (gene–culture loop). Herding cultures created an environment where adult milk digestion paid off, and lactase persistence spread in those populations in the last ~10,000 years. A cultural loop reached back into the genetic one. (Standard example; if used, cite Laland, Odling-Smee & Myles 2010, Nature Reviews Genetics, “How culture shaped the human genome”. Not verified here.)
Standing on shoulders (civilization, the ratchet named by its users). Newton wrote to Robert Hooke on 5 February 1675/76, “If I have seen further it is by standing on the shoulders of Giants.” He was drawing on an image John of Salisbury attributed to Bernard of Chartres in 1159. Scholars have been describing the ratchet for 850 years. Wikipedia: Standing on the shoulders of giants (modernized spelling; see Source material note)
Genome to vaccine inside a year (civilization, speed). The SARS-CoV-2 genome was posted publicly in January 2020. The Moderna vaccine’s phase 1 trial began in March 2020, and the first authorizations came in December 2020. Thousands of people across many countries ran the learning, creating and testing loops in parallel. Compare example 1. Wikipedia: COVID-19 vaccine; Moderna COVID-19 vaccine. (The exact genome posting date, around 10–11 January 2020 on virological.org, should be checked before use.)
Thought experiment: the island child. Picture a healthy newborn raised with food and shelter but no people, tools or words. At 30, what could they make? Now picture the same child raised in any village on Earth. The whole gap between the two is the loop, and none of it is in the child.
Everyday scene: a grandmother’s recipe card. A stained card with a crossed-out quantity and “less salt – Dad” in the margin. That one card holds all three loops: learning (someone tasted and measured), creating (someone changed the recipe), and becoming (the card is kept because of who wrote it). It has already passed through three generations. No other animal has a card like this.

Source material

Darwin, degree not kind. “Nevertheless the difference in mind between man and the higher animals, great as it is, certainly is one of degree and not of kind.” Charles Darwin, The Descent of Man, and Selection in Relation to Sex, 2nd ed. (1874), ch. IV. [verified: https://www.gutenberg.org/ebooks/2300] Use: This is the honest opening concession. The chapter should argue that scale and compounding make the difference, not a special faculty.
Darwin on language as a slow, undesigned loop. “no philologist now supposes that any language has been deliberately invented; it has been slowly and unconsciously developed by many steps.” Darwin, The Descent of Man, 2nd ed. (1874), ch. III. [verified: https://www.gutenberg.org/ebooks/2300] Use: Our most important loop, language, was not designed by anyone. This pairs with Kirby et al. (item 7).
The ratchet defined. “Human cultural transmission is thus characterized by the so-called ‘ratchet effect’, in which modifications and improvements stay in the population fairly readily (with relatively little loss or backward slippage) until further changes ratchet things up again.” Claudia Tennie, Josep Call & Michael Tomasello, “Ratcheting up the ratchet: on the evolution of cumulative culture,” Phil. Trans. R. Soc. B 364 (2009): 2405–2415, §1 (Introduction). [verified: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2865079/] Use: The chapter’s central mechanism. Tomasello’s Cultural Origins of Human Cognition (1999) is in references.json.
Why chimp culture plateaus. “Chimpanzee cultural traditions represent behavioural biases of different populations, all within the species’ existing cognitive repertoire (what we call the ‘zone of latent solutions’)” Tennie, Call & Tomasello (2009), abstract. [verified: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2865079/] Use: Gives the reader a precise name for what animal loops do and ours escape: culture that could be reinvented from scratch, versus culture no one could reinvent.
No one is smart enough alone. “humans may be smarter than other creatures, but none of us is nearly smart enough to acquire all of the information necessary to survive in any single habitat.” Robert Boyd, Peter J. Richerson & Joseph Henrich, “The cultural niche: Why social learning is essential for human adaptation,” PNAS 108, suppl. 2 (2011): 10918–10925, abstract. [verified: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3131818/] Use: Moves the “unnatural” from the individual brain to the collective loop. It pairs with the Franklin example and with Henrich’s The Secret of Our Success (2015), which is in references.json.
Whitehead on offloading thought. “Civilization advances by extending the number of important operations which we can perform without thinking about them.” Alfred North Whitehead, An Introduction to Mathematics (1911), ch. V “The Symbolism of Mathematics,” p. 61. [verified: https://archive.org/details/anintroductionto00whituoft] Use: Sophisticated loops compress hard-won learning into habits, tools and notation, which frees attention for the next turn. It links to Ch. 12 (mathematics) and Ch. 10 (thingification). Worth pairing with his next line about thought being “like cavalry charges”.
Design without a designer. “we provide the first experimental validation for the idea that cultural transmission can lead to the appearance of design without a designer.” Simon Kirby, Hannah Cornish & Kenny Smith, “Cumulative cultural evolution in the laboratory,” PNAS 105(31) (2008): 10681–10686, abstract. [verified: https://pubmed.ncbi.nlm.nih.gov/18667697/] Use: Shows the loop itself as the creative agent. It also cautions against crediting individual genius for what chains of people do.
Improvement without understanding. “These results indicate that complex technologies need not result from enhanced causal reasoning but, instead, can emerge from the accumulation of improvements made across generations.” Maxime Derex, Jean-François Bonnefon, Robert Boyd & Alex Mesoudi, “Causal understanding is not necessary for the improvement of culturally evolving technology,” Nature Human Behaviour 3 (2019): 446–452, abstract. [verified: https://pubmed.ncbi.nlm.nih.gov/30936426/] Use: A humbling point: our loops are smarter than the people running them. This connects to Ch. 21 (practices carry more than we understand).
Animal culture is real. “We find that 39 different behaviour patterns, including tool usage, grooming and courtship behaviours, are customary or habitual in some communities but are absent in others where ecological explanations have been discounted.” Andrew Whiten, Jane Goodall et al., “Cultures in chimpanzees,” Nature 399 (1999): 682–685, abstract. [verified: https://pubmed.ncbi.nlm.nih.gov/10385119/] Use: Keeps the chapter honest. Other animals have culture, and our difference is in how it compounds.
Culture across an ocean. “multiple song types spread rapidly and repeatedly in a unidirectional manner, like cultural ripples, eastward through the populations in the western and central South Pacific over an 11-year period.” Ellen C. Garland et al., “Dynamic horizontal cultural transmission of humpback whale song at the ocean basin scale,” Current Biology 21(8) (2011): 687–691, abstract. [verified: https://pubmed.ncbi.nlm.nih.gov/21497089/] Use: A strong counterexample to “only humans spread culture fast and far”. It sharpens the claim: whales’ songs turn over, while ours accumulate.

Also in references.json and worth mining (not quoted here): Jablonka & Lamb, Evolution in Four Dimensions (2005), on genetic, epigenetic, behavioural and symbolic inheritance. Maynard Smith & Szathmáry, The Major Transitions in Evolution (1995), which lists the origin of language as a major transition in how information is passed on. Odling-Smee, Laland & Feldman, Niche Construction (2003). Hutchins, Cognition in the Wild (1995), on navigation as cognition distributed across a crew and its instruments. Argyris & Schön, Organizational Learning (1978), on double-loop learning. Hofstadter, I Am a Strange Loop (2007). Stiegler on technics as exteriorized memory. Dittmar (2011) on the printing press.

Unverified quotation (do not use without checking): Newton to Hooke, 5 Feb 1675/76: “If I have seen further it is by standing on the shoulders of Giants.” [UNVERIFIED: checked only against Wikipedia’s modernized rendering. The manuscript spelling reportedly differs (“sholders”); check it against the Historical Society of Pennsylvania facsimile.]

Counterarguments and limits

Exceptionalism by the back door. Every species is “unnatural” at something: bats at echolocation, ants at agriculture, whales at cultural transmission over ocean basins. The chapter must say which measure makes us exceptional (compounding of retained modifications across generations and media) and admit that choosing that measure is itself a human cut (Ch. 8, Ch. 9).
“Degree, not kind” undercuts “unnatural”. If Darwin is right, why the word? A possible reply is that differences of degree compound. A ratchet that slips 10% per generation and one that slips 1% give utterly different worlds after 100 generations. The author should decide whether “unnatural” is a joke, a provocation, or a claim.
Cumulative culture may not need uniquely human minds. Caldwell & Millen (2009): imitation, emulation and teaching were “each… independently sufficient” for cumulative improvement, and “accounts of the unusual complexity of human culture in terms of species-unique learning mechanisms do not currently provide complete explanations” [verified: https://pubmed.ncbi.nlm.nih.gov/19891752/]. Demography, meaning population size and connectedness, may matter as much as cognition (Henrich 2004). So the chapter should not pin everything on the brain.
Some animals may ratchet a little. New Caledonian crow tool designs, and changes in chimpanzee and orangutan techniques, have been argued to show minimal cumulative culture. The claim should be “far more”, not “only”.
Sophisticated does not mean good. The loop carries errors, cruelty and myths as faithfully as truths (the hundredth-monkey myth itself is an example). Fast loops can destabilise: financial crises, misinformation cascades, climate change. Speed also outruns our ability to measure consequences (Ch. 18 runaway loops).
The ratchet is fragile. Tasmania, the Polar Inuit and the loss of Roman concrete techniques show that inheritance depends on living carriers and connected populations. The “across generations” claim is true on average and false in many particular places.
Becoming may not speed up. The third loop (marvel, meander, manifest) arguably resists acceleration. Ch. 16 refuses to justify it by productivity, and the author should not now smuggle productivity back in by praising our speed at it. It may be honest to say that culture multiplies the occasions for becoming (art, ritual, story) without making it faster.
Anthropocentric timelines. Comparing bacterial generations with human decades mixes units. Bacteria “learn” as populations very fast (CRISPR immunity, horizontal gene transfer, Ch. 19). Be precise about what is being compared: the loop’s cycle time, its fidelity, or its reach.
AI complicates the ending. Machine learning systems now run some learning loops faster than we do. That doesn’t refute the chapter, since these are loops we built, but a 2026 reader will ask about it. Dupoux, LeCun & Malik (2026) in references.json may help.

Connections

Ch. 6 (The Second Blade): It already promises that “we loop representation back into the real world faster and with more sophistication than any other life form”. This chapter pays that promise off.
Act I close (Unnatural in Volition) and Act II close (Unnatural in Distinction): Use a parallel structure so the three closes rhyme, and show the dependency chain: want, then cut, then loop.
Ch. 7 (The Third Blade): Ideas survive like organisms. The cultural ratchet is the machinery of that survival.
Ch. 10 (Thingification): Retained modifications harden into tools, institutions and “things”. Whitehead’s “without thinking about them” belongs here too.
Ch. 12 (Mathematics): Notation as the ultimate ratchet.
Ch. 13–16 (Learning, Creating, Becoming): The three loops being scaled up. Each gets one sentence of “here is what it looks like across a culture”.
Ch. 17 (Maintenance, Meaning, and Magic): The Polar Inuit case is a maintenance failure. Ratchets need upkeep.
Ch. 18 (Loops That Learn and Loops That Don’t): It supplies the runaway and stabilising mechanics. This chapter should not repeat them, only point back.
Ch. 19 (What Knowledge Needs to Stay Alive): Lateral transfer and word drift are the fine grain of the ratchet. The whale-song and Tasmania examples could go in either chapter, so decide where.
Ch. 20 (Raising the Floor): The collective loop is what raises the floor. Boyd et al.’s “none of us is nearly smart enough” is a moral point as well as a biological one.
Ch. 21 (Practices Before Patterns): Derex’s wheel shows practices improving without theory, which strongly supports Ch. 21’s humility.
Ch. 22 (A Better Tuesday): The reader’s own weekly loops (review date, stopping rule) are the smallest unit of this big claim.
Epilogue: It needs the one-line version of this chapter to fit into its single closing image.

Exercise ideas

Trace one object back. Pick an ordinary object within reach (a pencil, a spoon, a shoelace). Write down every step you would need in order to make it from scratch, alone, starting with nothing. When you hit a step you can’t do (mining graphite, smelting, spinning thread), write the name of the kind of person who can. Notice: how few steps you get through before you depend on strangers, and how many of those strangers are dead. Why: It makes the chapter’s core claim concrete in five minutes. The knowledge behind a simple object is spread across a culture and across centuries, in a way no animal’s world is.
Find the ratchet in your own skill. Choose something you do well (cooking a dish, a sport, a program you write, a language). List three things you do in it that you did not invent, and for each one, where you got it (a person, a book, a video, a correction). Then list one thing you added that someone else now does, or could. Notice: the ratio of inherited to invented, and whether your addition has somewhere to go (a person or record that will keep it). Why: It shows the reader that they are a link in a cumulative loop, both receiving and passing on. That is the difference between animal traditions and human ratchets. The last question also sets up the epilogue’s “build the world”.
The speed comparison. Pick one problem you solved in the last month (fixing a leak, learning a word, finding a route). Estimate how long it would have taken you with no other person, book or search, only trial and error. Then estimate how long a species would take to “solve” it by evolution, if it could at all. Notice: the orders of magnitude between the three answers, and which parts of the fast answer you could not personally explain (you trusted the loop). Why: It makes the “speed” half of the claim felt, and sets up the counterargument: trusting the loop is powerful and also how we get cheated. That ties back to the book’s goal.

Open questions for the author

What does “unnatural” mean here: a provocation, a joke, or a thesis? With Darwin’s “degree not kind” on the table, you need a clear line on whether compounding differences of degree amount to a difference in kind.
Which dimension carries the claim? Speed, fidelity (the ratchet), reach (across people), depth (loops about loops), or all four? A tight chapter might pick two.
Does “becoming” belong in the speed claim at all? Ch. 16 resists optimization. You could say culture multiplies becoming’s occasions rather than its speed, or quietly leave it out.
How much to concede to animals? Whale songs, chimp cultures and crow tools are strong material. Will they appear as honest concessions, as foils, or both?
Where does the Franklin/Polar Inuit material go? It fits here, Ch. 19 and Ch. 20. Choose one home and cross-reference from the others.
Should AI appear? A 2026 reader will think of it. Options are a sentence, a paragraph, or deliberate silence (with the epilogue picking it up).
Tone of the ending. Triumphant (“look how far we’ve come”), sober (“look how far our mistakes travel”), or both? The epilogue’s “use the blade carefully” suggests landing on responsibility.
How long should this close be? The act closes should probably match each other in length and structure. Decide the template (concession, difference, cost, hand-off?) once for all three.
Will you use the Newton line? If so, verify the original spelling against the manuscript, or use John of Salisbury’s 1159 version instead.

Reader perspectives

Curious young child

First reactions:

There’s nothing to read yet, only a brief. They’d say, “Where’s the story?”
“We’re faster than every animal? Cheetahs are faster.” They’ll take “speed” literally, so the chapter has to say it means learning speed, not running speed.
They’ll love the claim if it’s backed by animal comparisons, because animals are the best part of any book.
“Unnatural” makes them worry. “Is it bad to be unnatural? Like fake?”

Questions they’d ask:

“Do animals learn? My dog learned to sit.”
“Can crows really use tools?” (They’ve heard of this. New Caledonian crows making hook tools is a real, well-studied example.)
“Do bees teach other bees?” (The honeybee waggle dance, which tells hive-mates where food is, is real and would delight them.)
“If a chimpanzee learns something, does its baby know it too?”
“Why do humans have schools and animals don’t?”
“How did people learn so fast? Did someone start first?”
“If we’re so good at loops, why do grown-ups keep making the same mistakes?”
“Is being unnatural good or bad?”
“Does my great-great-grandpa’s learning live in me?”
“What’s the other two unnaturals again?” (They’ll have forgotten Acts I and II.)

Where they’d get lost, bored, offended or unconvinced:

“More sophistication and speed than any other life form” with no examples. They’d push back with the dog, the crow or the octopus.
“Across individuals, cultures and generations” is three abstract nouns. Each one needs a picture: me learning, my town learning, my family learning over time.
If the close sounds like bragging (“humans are the best”), a child who loves animals will be offended on their dog’s behalf.
Without a quick recap of the other two senses of unnatural, the “third sense” won’t mean anything to someone who forgot the first two.

Examples they’d bring:

Their dog learning tricks, and then the dog’s puppies not knowing any of them. That’s the gap humans cross with teaching.
Learning to read. Letters were invented by people long ago, and now a kid learns them in a year or two.
Grandma’s recipe book (from ch. 19). Learning that skips a generation’s worth of trying.
The Internet or YouTube: learning to do a yo-yo trick from a stranger across the world in ten minutes.
Bees dancing directions, as a comparison. It’s clever, but it’s the same dance for thousands of years.

What would win them over:

A side-by-side: what a crow or a bee can pass on, next to what a child passes on in one school year.
Saying clearly that “unnatural” doesn’t mean fake or bad. It means we do natural things so much that they turn into something new.
A three-line recap of all three unnaturals (volition, distinction, loops) that a kid could memorize.
Humility: “but we can also make loops that go wrong faster, too.”

Cynical adult

First reactions:

There’s no earlier draft. It’s an empty page with a brief, and the brief makes the biggest claim in Act III.
“We run these loops with far more sophistication and speed than any other life form” contradicts chapter 18’s draft, which warned that “higher” smuggles a verdict into a description. Which one is the book’s position?
“Speed” is simply wrong for some loops: bacteria evolve antibiotic resistance in days. Humans are fast at cultural loops, not all loops.
The “unnatural” framing is now on its third use. Unless this close earns it, it will sound like a tagline.

Questions they’d ask:

Faster and more sophisticated than which organisms, measured how?
Crows, chimps and whales have culture. What exactly do we do that they don’t, in a sentence?
Is “unnatural” a compliment, a warning or just a description here?
If our loops are so sophisticated, why do we keep repeating the same political, financial and personal mistakes?
Is speed good? Fast loops also produce bubbles, panics and viral lies.
“Across individuals, cultures and generations”: isn’t that just “we have writing”?
How is this different from the standard “humans are special because of culture” story in every popular anthropology book?
Does this close depend on the reader accepting MMM, or does it stand on its own?
What does being unnatural in our loops cost us? The book did that for the blade; will it do it here?

Where they’d get lost, bored, offended or unconvinced:

If this reads as a victory lap for humanity after chapters on humility, the tone breaks.
Any claim about “all life forms” without named comparisons will look like hand-waving.
If it summarizes chapters 13–22 again, I’ll skip it.
Leaning on “unnatural” as a punchline three times in one book may feel like branding rather than insight.

Examples they’d bring:

Whiten et al., “Cultures in chimpanzees,” Nature (1999): 39 behaviours that vary between chimp communities in ways not explained by ecology. Other animals have cultural loops; the human difference is one of degree and accumulation.
Michael Tomasello’s “ratchet effect” (The Cultural Origins of Human Cognition, 1999): human culture accumulates improvements without slipping back. This is a named, defensible version of the brief’s claim.
Antibiotic resistance: bacteria swap resistance genes laterally (a callback to chapter 19) faster than we develop drugs. That’s a loop where we’re losing the race.
The COVID-19 mRNA vaccines, developed within a year of the virus being sequenced (early 2020 to late 2020): the best recent example of human loops at speed, drawing on decades of slower, earlier work.
Financial flash crashes, such as the May 2010 US flash crash: our loops can now run faster than we can understand them.

What would win them over:

Replace “more sophisticated than any other life form” with a precise, defensible claim: cumulative, cross-generational and symbolic loops, with the ratchet named.
Show one cost alongside the gift, matching the “cost of the blade” move from Act II.
Keep it short, one image rather than a recap.

Believer / spiritual reader

First reactions:

The claim that we run our loops “with far more sophistication and speed than any other life form” will sound to a Jewish or Christian reader like Psalm 8 without God: “What is man, that thou art mindful of him?… For thou hast made him a little lower than the angels” (Psalm 8:4–5, KJV). They’ll want to know whether the book is making the same claim on secular grounds, and whether it knows it.
Other believers will hear it as a boast, and after twenty chapters of humility, it’s a risky note. Pope Francis’s Laudato si’ (2015) criticizes exactly this “technocratic paradigm” and misplaced anthropocentrism. Buddhist and Jain readers will ask why sophistication is the measure at all.
There is no draft yet, only the brief. A believer can’t judge the tone, but will be watching whether “unnatural” here means proud or burdened.
They’d be moved if the close admits that our fast loops also give us faster ways to go wrong: sin at scale, in religious terms.

Questions they’d ask:

Is being unnatural in our loops a gift, a burden or a danger? Genesis gives humans dominion (1:28) and also the job to “dress it and to keep it” (2:15). Which reading does the book lean toward?
More sophisticated than other life forms at what? A Jain would say a monk who harms nothing is more advanced than an engineer who builds faster.
If our loops run across generations, doesn’t that make tradition, including religious tradition, the main engine of our unnaturalness? Will the close say so?
Is speed good? The contemplative traditions have mostly been efforts to slow the loops down.
Does “unnatural” mean we stand outside nature, or that we are a strange part of it? Some traditions (stewardship) need the first, some (Indigenous kinship, Daoism) resist it.
If a whale’s song changes across a whole ocean population over a season, or crows teach each other faces, how unnatural are we really? Is this a difference of degree or kind?
How does this close connect to the other two (volition, distinction)? Religious readers will hear free will, the knowledge of good and evil, and the long human story, which is the Eden narrative in three acts.
Is the book going to say why having sophisticated loops matters, or only that we have them?

Where they’d get lost, bored, offended or unconvinced:

Offended: Any hint of triumph (humans as the peak of life) without the cost. Religious ecology movements, and many Hindu, Buddhist and Indigenous readers, will reject a human-exceptionalist close.
Unconvinced: “Far more sophistication” without evidence. The Chapter 18 draft says “life does not climb toward complexity as a general destination” and “‘Higher’ smuggles a verdict into a description.” The close must not contradict that.
Lost: If the close lists learning, creating and becoming again without adding anything, it will read as a summary.

Examples they’d bring:

Psalm 8:3–5: a human looking at the heavens, feeling small, and being told they have been given great standing anyway. It’s the exact emotional shape the close is reaching for: tiny and unnatural at the same time.
Genesis 11, the Tower of Babel: the story of a people who coordinate so well that “nothing will be restrained from them,” and whose coordination is broken up. It’s an ancient warning about how powerful shared loops are.
Pope Francis, Laudato si’ (2015), chapter 3: a critique of the “technocratic paradigm” and of an anthropocentrism that treats nature as raw material. It gives the close a serious religious voice warning about the same unnaturalness it describes.
The Jain principle of ahiṃsā (non-harm): a tradition that measures a life by how lightly it loops back on the world, not how powerfully. A counterweight to sophistication as the measure.
Robin Wall Kimmerer, Braiding Sweetgrass (2013): a botanist and citizen of the Potawatomi Nation writing about reciprocity and the “Honorable Harvest.” It offers a view where humans are younger siblings among other beings, not the most sophisticated one.

What would win them over:

Framing our unnatural loops as responsibility as much as power, with the cost named in the same breath.
Keeping Chapter 18’s warning against “higher” in the close itself.
Admitting openly that traditions have said something like this before (Psalm 8, Babel, Laudato si’), and that the book is joining an old conversation.

Skeptical scientist

First reactions:

No draft yet, only the brief. That’s an opportunity to get the central claim right from the start, because as written it’s overstated.
“Far more sophistication and speed than any other life form” is false for speed in the obvious sense. Bacteria evolve adaptations in days; E. coli can double in about 20 minutes under good lab conditions. Humans win on something else: cumulative culture, where improvements stack across generations without being reinvented. Name that and the claim becomes defensible.
Other animals have learning, culture and traditions. The honest claim isn’t “only we loop” but “only we ratchet this far, this fast, across this many domains at once.”
This is the right place to show gene-culture coevolution: human loops are fast enough to change our own biology.

Questions they’d ask:

Faster than what, measured how? Per generation, per year, per individual lifetime?
What’s “sophistication” here: number of nested levels, fidelity of copying, amount accumulated, range of domains? Pick one you can defend.
How do you handle bacterial CRISPR, which is literally a heritable memory of past infections?
How do you handle the vertebrate adaptive immune system, which runs variation and selection inside one body in days?
Which non-human animals have culture, and what exactly do they lack?
Is the difference in kind or in degree? If degree, why call it “unnatural”?
Does the evidence say cumulative culture is uniquely human, or is that contested?
Is there a case where human cultural loops changed human genes?
Are there human loops that are slower or worse than other species’? (We are terrible at some kinds of learning compared with specialists.)
Does speed matter if the loop can also lose knowledge fast?

Where they’d get lost, bored, offended or unconvinced:

Unconvinced by any claim of superiority phrased as “any other life form” without qualification. A microbiologist or immunologist would stop reading.
Offended on behalf of animal-culture researchers if the chapter implies culture is human-only.
Bored by a triumphant tone. The brief sets up a wonder moment; it should rest on specific evidence, not rhetoric.

Examples they’d bring:

Bacterial learning loops. Barrangou et al. (2007, Science) showed that bacteria acquire resistance to phages by inserting fragments of phage DNA into their CRISPR loci, a heritable record of past infections. It’s a loop that learns, and it predates us by billions of years. Concede it, then locate the human difference.
Evolution within a body. Adaptive immunity uses somatic hypermutation and clonal selection to produce high-affinity antibodies within days to weeks. Variation plus selection, within one organism’s lifetime. Another loop that isn’t slow.
Animal culture. Whiten et al. (1999, Nature, “Cultures in chimpanzees”) catalogued behavioral traditions differing between chimpanzee communities. Garland et al. (2011, Current Biology) documented humpback whale songs spreading eastward across the South Pacific population by population, a cultural wave across an ocean basin.
The ratchet. Tomasello’s “ratchet effect” (The Cultural Origins of Human Cognition, 1999) and Tennie, Call and Tomasello (2009, Philosophical Transactions of the Royal Society B, “Ratcheting up the ratchet”) argue that high-fidelity transmission lets human culture accumulate improvements that no individual could invent alone. This is the defensible version of the brief’s claim.
Culture changing genes. Lactase persistence into adulthood spread in dairying populations. Tishkoff et al. (2007, Nature Genetics) showed independent (convergent) mutations for lactase persistence in East African pastoralists and Europeans. Human cultural loops were fast and strong enough to drive genetic selection.
Demography matters. Kline and Boyd (2010, Proceedings of the Royal Society B) found that toolkit complexity in Oceanic island societies correlated with population size and contact. Our loops are fast because there are many of us connected; that’s a population property, not only an individual one.

What would win them over:

A narrower, defensible claim: cumulative, high-fidelity, multi-domain culture across generations, fast enough to reshape our own genome and planet.
A fair acknowledgment of CRISPR, adaptive immunity and animal culture before stating what’s distinctive.
Using “degree” honestly: say whether “unnatural” means a difference in kind, or a difference in degree so large it behaves like one.